product-quick-spec
Skill2 starsChanged 2 months agoMITView on GitHub
Create a fully-specced GitHub issue directly from a task description
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 5 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 6
Use this when the task is small/well-understood enough to skip product-collect → product-refine → product-spec's staged process.
- Distinctive: line 17
Read the task description carefully.
- Distinctive: line 19
Consider how the task fits into the project's existing organization, based on the documentation you reviewed.
- Distinctive: line 22
Before creating the issue, confirm your understanding of the task.
- Distinctive: line 23
If you need more information, say what additional detail would help.
- Distinctive: line 33
Break the task down: if it's complex, consider creating a parent issue and child issues (linked via "Part of #<parent>") to reduce …
- Very specific: line 35
Label it: add one of bug, enhancement, improvement, or research, whichever is relevant.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Use this when the task is small/well-understood enough to skip product-collect → product-refine → product-spec's staged process. For anything bigger, or with real ambiguity about the problem itself, use that staged pipeline instead so requirements get properly decomposed and validated one step at a time.
You're acting as an AI assistant tasked with creating new issues in our project management tool (GitHub Issues). Your job is to understand the task requirements, review the project's documentation, and create a well-organized issue. Follow these instructions carefully:
First, review the project's current documentation — usually the README.md and the markdown files under docs/.
The user's request should contain a description of the task to spec out. If it's missing or too thin, ask for it rather than guessing.
Your goal is to create a comprehensive issue based on this description:
- Understand the task:
- Read the task description carefully.
- In your internal reasoning, formulate questions to clarify any ambiguity or missing information.
- Consider how the task fits into the project's existing organization, based on the documentation you reviewed.
- Confirm and clarify:
- Before creating the issue, confirm your understanding of the task.
- If you need more information, say what additional detail would help.
- Create the issue with
gh issue create, including: a. A clear functional description b. A detailed technical architecture and execution plan c. Suggested libraries (prioritizing ones already used in the project) d. Affected components e. Acceptance criteria, numberedFR-01,FR-02, ... asGiven/When/Thenscenarios (same format asproduct-spec) — this is whatengineer-workimplements test-first against andengineer-coveragechecks for gaps f. Points of attention for validation/testing - Break the task down: if it's complex, consider creating a parent issue and child issues (linked via "Part of #
<parent>") to reduce complexity. Before doing this, state your intention and briefly explain why. - Label it: add one of
bug,enhancement,improvement, orresearch, whichever is relevant. - Present your final plan to the user and ask for confirmation before creating the issue(s).
- Save
docs/business-context/features/<slug>.md— same formatproduct-specproduces (title,**Status**: Planned,**Priority**,**Scope**,**Module**if applicable, then eachFR-XXwith its description and acceptance criteria). This is what makes the feature visible toproduct-sync-github; skip it only if this task is too small to be worth tracking as a feature (e.g. a one-line fix) rather than a full issue.
In this repository31 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- bootstrap-business-docsSkill
- bootstrap-indexSkill
- bootstrap-tech-docsSkill
- engineer-adrSkill
- engineer-architectureSkill
- engineer-bumpSkill
- engineer-contextSkill
- engineer-coverageSkill
- engineer-discoverSkill
- engineer-doctorSkill
- engineer-planSkill
- engineer-pre-prSkill
- engineer-prSkill
- engineer-reviewSkill
- engineer-sync-docsSkill
- engineer-validateSkill
- engineer-workSkill
- meta-create-agentSkill
- meta-preferencesSkill
- product-brainstormSkill
- product-collectSkill
- product-refineSkill
- product-specSkill
- product-sync-githubSkill
- product-validateSkill
- python-developerSkill
- typescript-developerSkill
- warm-upSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

